Llama 4 Maverick is a 400 billion parameter Multimodal LLM (with 17 billion active parameters and 128 experts) designed for advanced multilingual, multimodal reasoning and enterprise-scale applications. It excels in high-quality text and image understanding, creative writing, coding assistance, long-context tasks (up to one million tokens), and robust multilingual support across twelve major languages[1][6][8].

Some other noteworthy features of Llama 4 Maverick include native multimodality—allowing seamless integration of text and images in both input and output—and efficient inference via its sparse Mixture-of-Experts architecture. This enables scalable deployment with lower latency while maintaining strong performance on complex reasoning, chat, document intelligence (including diagrams/tables), customer support with visual context, content generation from PDFs or images, and more.

Metric	Value
Parameter Count	400 billion total / 17 billion active
Mixture of Experts	Yes
Active Parameter Count	17 billion
Context Length	Up to 1 million tokens (524k currently on Together.ai; roadmap to full support)
Multilingual	Yes
Quantized*	Yes
Precision*	FP8

*Quantization is specific to the inference provider and the model may be offered with different quantization levels by other providers.